Androgen ablation is the only form of systemic therapy demonstrated to prolong life in men with advanced prostate cancer (CAP), precipitating apoptosis in subpopulations of cells. Despite high initial response rates, remissions are temporary because surviving CaP cells usually recur with an androgen independent (AI) phenotype. Thus, one of the main obstacles to a cure of advanced CaP by androgen ablation is AI progression, a complex process involving variable combinations of clonal selection, adaptive upregulation of cell survival genes, and ligand-independent androgen receptor or alternative growth factor activation. We hypothesize that changes in apoptosisassociated gene expression play critical roles in tumor progression and drug resistance and that adaptive changes in gene expression after androgen withdrawal increase CaP cell survival, accelerate AI progression, and render cells chemoresistant. Identification and functional characterization of molecular events that allow cancer cells to survive treatments that normally induce apoptosis will improve our understanding of progression and identify new therapeutic targets. Application of microarray gene expression analysis to complimentary Shionogi and LNCaP model systems after apoptotic triggers like castration or chemotherapy will serve as a primary screen to identify apoptosisassociated gene clusters. Advantages of determining gene expression profiles in these well-defined xenograft models, compared to direct hybridization of human CaP tissues, include their predictability and reproducibility, circumventing inherent problems of small quantity of cancer typically present in regressed CaP specimens after androgen withdrawal and the heterogeneity of specimens from individual patients. Apoptosis-associated genes will undergo secondary screening in human CaP tissue arrays (comprised of cores from hormone naive; 1, 2, 3, and 8 months after neoadjuvant hormone therapy (NHT); androgen independent; and post NHT + Taxotere) to verify relevance to the human disease. Candidate genes with putative pro- or anti-apoptotic functions will then be subjected to tertiary analysis for functional characterization in xenograft models (i.e., overexpression vs antisense-inhibition) or transgenic knockouts to clarify role in apoptosis and hormone resistance. As examples, in this proposal we plan to extend on our previous experience with Bcl-2, clusterin, IGFBP-2 and IGFBP-5 genes and characterize functional roles of 2 apoptosis-associated genes that increase after castration in Shionogi tumors - Hsp27 and p35 (regulator of cdk5). Finally, as an illustrative example of quaternary analysis, we will initiate a clinical trial using ASO's to target clusterin, an anti-apoptosis gene that we have recently shown in primary, secondary, and tertiary screens to increase after cell death triggers and enhance cell survival. We will evaluate the safety and activity of a 2nd generation antisense oligonucleotide targeting clusterin in a Phase I/II presurgical pharmacokinetic and pharmacodynamic study in men with localized prostate cancer. Collectively, these objectives will improve our understanding of apoptosis, adaptive cell survival mechanisms, drug resistance, and identify new therapeutic targets for clinical development.